Github user sureshsubbiah commented on a diff in the pull request:
https://github.com/apache/incubator-trafodion/pull/44#discussion_r36356098
--- Diff: core/sql/executor/ExHbaseAccess.cpp ---
@@ -1645,12 +1650,18 @@ Lng32
ExHbaseAccessTcb::createSQRowFromAlignedFormat(Int64 *latestRowTimestamp)
if (convertExpr())
{
+ convertRowLen_ = hbaseAccessTdb().convertRowLen();
+ UInt32 rowLen = convertRowLen_;
ex_expr::exp_return_type evalRetCode =
- convertExpr()->eval(pentry_down->getAtp(), workAtp_, NULL, asciiRowLen);
+ convertExpr()->eval(pentry_down->getAtp(), workAtp_, NULL,
+ asciiRowLen, &rowLen);
if (evalRetCode == ex_expr::EXPR_ERROR)
{
return -1;
}
+ if (hbaseAccessTdb().getUseCif() &&
+ rowLen < convertRowLen_)
+ convertRowLen_=rowLen;
--- End diff --
Yes rowLen can be hbaseAccessTdb().convertRowLen(). Row buffer has been
pre-allocated with hbaseAccessTdb().convertRowLen(). Now that we have set
convertRowLen_ to be actual size of output row for this particular row, I think
this ensure that we copy only the required bytes for this output row. Please
see change ExHbaseSelect.cpp
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
